We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the "normal sound" from observation of the microphone, and then detects sounds never observed before as "abnormal sounds." To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection. As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.